-
When this checklist applies
-
The seven-step spec check
-
Step 1: Photograph the data plate—don't trust the model number you remember
-
Step 2: Identify which filter you're actually replacing
-
Step 3: Read the micron rating and separation efficiency together
-
Step 4: Confirm rated flow at your actual operating pressure
-
Step 5: Measure the physical envelope and connection—not just the part number
-
Step 6: Verify temperature and differential-pressure limits
-
Step 7: Confirm stock and lead time before you commit
-
Step 1: Photograph the data plate—don't trust the model number you remember
-
Notes and common mistakes
When this checklist applies
This is for the call where a plant manager tells you the Atlas Copco compressor is in alarm, the element is clogged, and the line stops Friday if the replacement doesn't show up. I've coordinated a couple hundred emergency filter and spare-part orders over the past six years—same-day turnarounds included—and the pattern is always the same: the delay almost never comes from shipping. It comes from ordering the wrong spec.
Seven steps below. Run them in order. The one most people skip is step five, and that's usually the one that triggers the reorder.
The seven-step spec check
Step 1: Photograph the data plate—don't trust the model number you remember
People call and say "we've got a GA 37," and that's fine as a starting point, but the filter housing model is a separate identifier. It can differ by build year and by configuration even within the same compressor family. Before you call a supplier, get a photo of the machine data plate: model, serial number, year of manufacture. If there's a label on the filter housing itself, photograph that too.
We had two units on the same floor—same nominal model, different filter housings because one was a 2019 build and the other was a 2021. I want to say the split was around 2020, but don't quote me on the exact year; check the plate. Last quarter we dodged a bullet when the photo showed a housing series the caller had assumed was something else entirely. That one photo saved a two-day reorder.
Step 2: Identify which filter you're actually replacing
"Air filter" gets used loosely. On a typical oil-injected screw compressor you're dealing with the intake air filter, the oil filter, the oil/air separator, and possibly an inline coalescing filter downstream. Four different elements, four different spec sheets, four different part numbers. In a rush, people conflate "the thing that's beeping" with "the air filter," and the order goes sideways.
Checkpoint: match the alarm code or the maintenance log entry to the physical location on the machine. If you can't, ask the technician to send a photo of the element they pulled out. That sounds obvious. It isn't, at 4 p.m. on a Thursday.
Step 3: Read the micron rating and separation efficiency together
A micron number on its own doesn't tell you much. A 3-micron element rated at 99.9% separation is a different product from a 3-micron element rated at 95%. If you're matching to a compressed-air quality requirement, work from the ISO 8573-1 class your process needs—that's the standard most plants are actually trying to hit, whether they say so or not—and then check the element's stated efficiency curve against it.
This was accurate as of January 2025 when I last pulled datasheets. Media formulations get revised, so verify the current efficiency data with the supplier before you commit to a spec. I've seen a "same part, new media" revision change the efficiency number by a couple of points without the part number changing.
Step 4: Confirm rated flow at your actual operating pressure
Rated flow is usually quoted at a specific reference pressure—often 7 bar, sometimes 8. If your site runs at 10 bar or at 4 bar, the effective capacity and pressure drop across the element change. An element that's correctly sized on paper at the reference pressure may be undersized at your actual conditions.
Checkpoint: put your site's normal operating pressure next to the rating pressure on the datasheet. If they don't match, ask the supplier to confirm the corrected flow. This is a five-minute question that prevents a filter that clogs in three weeks instead of six months.
Step 5: Measure the physical envelope and connection—not just the part number
Here's the one people skip. Even with a matching part number, housings get revised, and connection details can shift between production batches. Before you order, measure the element: outside diameter, inside diameter, overall height, and note whether the connection is threaded, flanged, or a push-fit. If there's an O-ring or gasket, note the profile.
I have mixed feelings about how much measurement is enough. On one hand, it feels excessive when the part number matches and the supplier confirms fitment. On the other, the two emergency reorders I remember most clearly both happened because someone trusted the number and didn't measure. So now our rule is simple: photo, tape measure, supplier confirmation, then purchase order. In that order.
Step 6: Verify temperature and differential-pressure limits
Ambient temperature, altitude, and duty cycle all affect element life and the differential-pressure alarm setpoint. If you're in a hot enclosure or at elevation, the element's rated limits may need derating. Check that the DP alarm setpoint on the machine matches the element's rating—mismatched setpoints are a quiet cause of premature changeouts.
The "change it when it looks dirty" approach comes from an era when DP gauges weren't standard. Today the gauge and the alarm log give you better data. Use them.
Step 7: Confirm stock and lead time before you commit
"In stock" and "listed as available" are not the same thing. Ask which warehouse the part sits in, and ask what the cutoff time is for same-day dispatch. If you're sourcing through a regional entity—PT Atlas Copco in Indonesia, for example—lead time depends on whether the element is in the local warehouse or has to come from a regional hub. That's a different clock, and it's worth confirming in writing.
This is where an efficient process pays off. If you keep a standard record of the data from steps one through six for each machine, the next reorder takes minutes instead of an afternoon. Reducing the manual back-and-forth reduces the errors. That's been the clearest efficiency win in our maintenance workflow, and it didn't require new software—just a consistent form.
Notes and common mistakes
The most expensive mistake is the "if it fits, it works" logic. It fits because the housing dimensions are similar, not because the filtration performance matches your process requirement. Fit is necessary; it isn't sufficient.
On aftermarket or third-party elements, I'll be honest: I have mixed feelings. The price gap is real, and some third-party suppliers hold legitimate certifications. But I've also seen efficiency claims that didn't hold up against the datasheet. If you go that route, ask for the test standard and the efficiency curve, not just the micron number. That's a judgment call, and it depends on how critical the air quality is for your process.
One more: don't wait for the alarm. Log the differential-pressure trend weekly. A rising trend tells you weeks in advance when the element is due, which turns an emergency into a scheduled order. That single habit has cut our rush filter orders by more than half.
This worked for our operation, but we run three shifts with fairly predictable load. If you're a seasonal plant with demand spikes or a site with highly variable ambient conditions, the changeout intervals and the urgency calculus might look different. I can only speak to the mid-size industrial context I've worked in. Verify current specifications and lead times with your supplier before you place a rush order—the market moves, and so do part revisions.